A comprehensive assessment of precipitation at Sierra de Guadarrama through observation and modeling

Abstract

The main objective of this PhD Thesis is to make a high resolution assessment of the climatological behavior and variability of precipitation at Guadarrama based on measurement and modeling. The methods used to achieve this are: design, installation and operation of a mountain meteorological network; climatological analysis of the historical precipitation observations available in Sierra de Guadarrama; analysis of the relationships between synoptic forcing and precipitation in Sierra de Guadarrama using Principal Component Analysis and k-means for weather typing and physical modeling of the precipitation processes using a orographic precipitation model with very high resolution (200 meters). The results are a dataset of high altitude meteorological observations from 1999 to 2014 at five locations in Sierra de Guadarrama, which enables to have a first estimation of the climate of this region and an unique opportunity to acquire knowhow on meteorological observing techniques and operational practices at mountains. The limitation on having reliable measurements of precipitation at this mountain area, led to to perform an analysis of the historical records of observations made at three AEMET observatories for the period 1989-2012. Some of the features of precipitation found are the pronounced summer drought, the high inter-annual variability and the strong connection with the advection of humidity from the Atlantic Ocean, strong altitudinal gradient, the complexity of the spatial patterns. The strong connection found between the advection of humidity from the Atlantic and local precipitation led to perform an analysis of the connections between the advection of humidity and precipitation at Guadarrama using weather typing methodologies. This analysis stressed the importance of orographic precipitation as the phenomena explaining the altitudinal gradient. In order to confirm the hypothesis of orographic precipitation being a strong factor, the Linear Model of Orographic Precipitation (Smith and Barstad, 2004) was applied for simulating precipitation at Guadarrama from 1990 to 2013 with 200 m resolution and a time step of 6 hours. Results show a good agreement with observations at all scales, specially at monthly scale. Thanks to the application of this model to Guadarrama, a comprehensive assessment of precipitation through observation and modeling is now available for the community. The gridded data base of precipitation now available going from 1990 to 2013, with 200 m spatial resolution and daily temporal resolution opens new possibilities for research at this area.